Are AI scams and cyberattacks rising in Hong Kong? What the data really shows
Yes for overall cyber incidents: HKCERT says Hong Kong security incidents hit a record high, up 27% year on year, but that is not a standalone AI scam growth rate.[5] AI enabled fraud is no longer hypothetical: HKCERT flags AI related attacks and AI content hijacking, and Hong Kong has seen a deepfake video call sca...
Published byEdited with GPT-5.5Images generated with GPT Image 2
Yes for overall cyber incidents: HKCERT says Hong Kong security incidents hit a record high, up 27% year on year, but that is not a standalone AI scam growth rate.[5]
AI enabled fraud is no longer hypothetical: HKCERT flags AI related attacks and AI content hijacking, and Hong Kong has seen a deepfake video call scam involving more than US$25 million, about HK$200 million.[5][3][4][6]
The safest response is procedural: verify high value transfers through independent channels, treat urgent payment requests as high risk, and assign clear cybersecurity ownership inside companies.[4][6][5]
香港 AI 詐騙同網絡攻擊係咪多咗？HKCERT 數據與深偽案例查證AI 生成編輯插圖：香港網絡安全與深偽詐騙風險。
AI Prompt
Create a landscape editorial hero image for this Studio Global article: 香港 AI 詐騙同網絡攻擊係咪多咗？HKCERT 數據與深偽案例查證. Article summary: 係：以 HKCERT 保安事故計，香港網絡安全事故已創新高、按年升 27%；但公開官方資料未足以精確回答「AI 詐騙案件升咗幾多」。[5]. Topic tags: hong kong, cybersecurity, ai, deepfakes, scams. Reference image context from search candidates: Reference image 1: visual subject "# HKCERT Releases “Hong Kong Cybersecurity Outlook 2026” Security Incidents Hit Record High with 27% Annual Increase AI-related Attacks and Supply Chain Risks Emerge as Top Concern" source context "HKCERT Releases “Hong Kong Cybersecurity Outlook 2026” Security Incidents Hit Record High with 27% Annual Increase AI-re" Reference image 2: visual subject "根據報告，2025年香港共錄得15,877宗網絡安全事故，相比2024年按年激增27%，創下歷年新高。HKCERT亦同日發表《香港企業網絡安全現況》研究，分析本地企業的" source context "Menu Open" Style: premium digital editorial illustration, source-backed research mood, clean composition
openai.com
Before reading too much into dramatic AI-fraud headlines, separate two questions. One is whether Hong Kong’s recorded cyber-security incidents are rising. The other is whether AI or deepfake scams, specifically, have an official year-on-year growth figure. The evidence is strong on the first point and more limited on the second: public sources show a worsening cyber-risk environment and serious AI-enabled fraud cases, but they do not yet provide a clean official time series for AI scams alone.
Studio Global AI
Continue your research
This page includes a source-backed answer you can continue inside Studio Global.
What is the short answer to "Are AI scams and cyberattacks rising in Hong Kong? What the data really shows"?
Yes for overall cyber incidents: HKCERT says Hong Kong security incidents hit a record high, up 27% year on year, but that is not a standalone AI scam growth rate.[5]
What are the key points to validate first?
Yes for overall cyber incidents: HKCERT says Hong Kong security incidents hit a record high, up 27% year on year, but that is not a standalone AI scam growth rate.[5] AI enabled fraud is no longer hypothetical: HKCERT flags AI related attacks and AI content hijacking, and Hong Kong has seen a deepfake video call scam involving more than US$25 million, about HK$200 million.[5][3][4][6]
What should I do next in practice?
The safest response is procedural: verify high value transfers through independent channels, treat urgent payment requests as high risk, and assign clear cybersecurity ownership inside companies.[4][6][5]
Yes, if you use HKCERT’s recorded security-incident data: its 2026 outlook says incidents hit a record high and rose 27% year on year.
Is AI now a recognised cyber-risk category?
Yes. HKCERT’s 2026 outlook names AI-related attacks among top concerns, while its 2025 outlook lists AI content hijacking as a key emerging risk.
Has Hong Kong seen real deepfake fraud?
Yes. A Hong Kong finance worker was reportedly tricked into transferring more than US$25 million after fraudsters used deepfake technology to pose as colleagues on a video call; the AI Incident Database describes a Hong Kong case involving about HK$200 million.
Can we say official data proves AI scams rose by 1,000%?
Not from the public sources here. A media report citing Sumsub says Hong Kong deepfake incidents rose 1,000% in Q1 2024, but that is not the same as a complete official Hong Kong time series for AI-scam cases.
What the official data actually says
HKCERT — the Hong Kong Computer Emergency Response Team Coordination Centre — is the key source for the clearest number. In its Hong Kong Cybersecurity Outlook 2026, HKCERT says security incidents in Hong Kong reached a record high, rising 27% year on year. The same release identifies AI-related attacks and supply-chain risks as top concerns, and says nearly 30% of enterprises lack dedicated cybersecurity personnel.
That 27% figure matters, but it has to be used carefully. It refers to overall security incidents, not a separate official count of AI scams. In other words, it is fair to say recorded cyber-security incidents are up; it is not fair to rewrite that number as ‘AI scams rose 27%’.
The previous year’s HKCERT outlook also pointed in the same direction. Its 2025 briefing said phishing had reached a five-year high, and highlighted supply-chain vulnerabilities and AI content hijacking as major emerging risks. The Hong Kong Police Force’s Cybersecurity Report 2024, produced by its Cyber Security and Technology Crime Bureau, covers the global and local cyber-security landscape, offers analysis and recommendations, and forecasts threats facing Hong Kong, reflecting continued official attention to defensive readiness.
AI-enabled fraud is no longer theoretical
The strongest Hong Kong example is the deepfake video-call fraud reported in 2024. CNBC reported that a Hong Kong finance worker was persuaded to transfer more than US$25 million after fraudsters used deepfake technology to disguise themselves as colleagues in a video call.
The AI Incident Database records a Hong Kong case in which attackers allegedly simulated a CFO and other video-conference participants, persuading an employee to transfer about HK$200 million, or roughly US$25 million.
The lesson is not just the size of the loss. It is the change in the trust model. Seeing a familiar face on a call, hearing a familiar voice, or receiving a convincing chat message can no longer be enough to approve a sensitive action, especially a high-value payment.
How to read the ‘1,000% deepfake increase’ claim
A FutureCIO article citing Sumsub data says Hong Kong saw a 1,000% increase in deepfake incidents in the first quarter of 2024. That is a useful warning sign, but it should not be treated as interchangeable with HKCERT or Hong Kong Police Force whole-market statistics.
The cautious reading is: commercial and media-reported data support the view that deepfake risk is escalating, while the public official evidence supports a broader rise in cyber-security incidents and heightened AI-related concern. What the available sources do not establish is an official, comparable, year-by-year percentage increase for AI-scam cases alone.
What people and companies should do now
Do not approve large transfers on the strength of a video call, email or chat message alone. The Hong Kong deepfake case worked by making a fake video meeting appear credible. Use a separate channel — for example, a known phone number, an internal workflow, or a second authorised approver — before releasing money.
Make urgency a trigger for extra checks. Requests framed as confidential, time-critical, or coming from a senior executive should not bypass payment controls. Deepfake fraud is dangerous because it can turn a familiar face or voice into pressure; the safeguard should be a process, not a gut feeling.
Keep basic phishing defences sharp. AI may be the new headline, but phishing remains a central risk. HKCERT’s 2025 outlook said phishing had reached a five-year high, so staff training, URL checking, login-page scrutiny and cautious handling of attachments still matter.
Assign clear cybersecurity ownership. HKCERT’s 2026 outlook says nearly 30% of enterprises lack dedicated cybersecurity personnel. Even a small company should know who owns incident response, payment verification rules, supplier access reviews and escalation procedures.
Review suppliers and third-party access. HKCERT’s 2026 outlook names supply-chain risk as a top concern, and its 2025 outlook also flags supply-chain vulnerabilities as an emerging risk. Vendor accounts, shared logins, cloud permissions and outsourced systems should be reviewed regularly.
Bottom line
If the question is whether Hong Kong’s recorded cyber-security incidents have increased, the answer is yes: HKCERT says they hit a record high and rose 27% year on year.
If the question is exactly how much AI-scam cases have increased, the public evidence is less precise. AI-related attacks, AI content hijacking and deepfake video-call fraud are real and serious risks in Hong Kong, but a single official percentage increase for AI scams alone is not established by the sources available here.
hkcert.orgHKCERT Releases “Hong Kong Cybersecurity Outlook 2026” Security Incidents Hit Record High with 27% Annual Increase AI-related Attacks and Supply Chain Risks Emerge as Top Concerns Nearly 30% of Enterprises Lack Dedicated Cybersecurity Personnel
cnbc.comDeepfake scams have looted millions. Experts warn it could get worse